Google told employees on Friday that it was closing its investigation into the expulsion of leading AI researcher Timnit Gebru. The company declined to say what the internal investigation found, but said it was making some changes in addressing research, diversity and employee issues.
Why it matters: The treatment of Gebru, before and after she was forced out of the company, angered people in Google’s Ethical AI team and others inside and outside the company. Google’s handling of the case has also raised questions about the company’s commitment to diversity and the use of ethicists who are free to question the company’s actions.
Under its new policies, Google states that it:
- bind the pay for those at the Vice Presidential level, and in part to achieve diversity and inclusion goals.
- streamlines the process of publishing research.
- increase staff related to employee retention.
- introduces new procedures around potentially sensitive employee exits.
“I understand we could and should have handled this situation with greater sensitivity,” Google CEO Jeff AI said in a memorandum obtained by Axios on Friday, setting out the changes. “And I’m sorry for that.”
I have heard and acknowledged what Dr. Gebru’s retirement means for female technologists, for those in the Black community and other under-represented groups pursuing their careers in technology, and for many people who are very interested in Google’s responsible use of AI. This has led some to question their place here, which I regret.
– Jeff Dean, head of Google AI, in an internal email on Friday
Gebru is a leading AI ethics researcher and one of a handful of prominent black women in the field, and Google has regularly promoted her work as a way to clarify her commitment to the AI ethics field. But the relationship deteriorated as her research became critical of the company’s work, which eventually led to her forced departure.
Between the lines: None of these moves are likely to quell the ongoing frustration over this issue, especially not within Google’s Ethical AI team. remain blinded by the movements of the company.
- Earlier this week, Google appointed Marian Croak as head of ethical AI work across Google’s research organization. Croak is a longtime employee of Google, a black woman and a prominent researcher who has done pioneering work in the field of dealing with voice via the Internet; however, none of her work is directly related to ethical AI.
- Her appointment was first reported by Bloomberg News on Wednesday night before an announcement to staff the next day.
Meanwhile, one of Gebru’s colleagues, Margaret Mitchell, remains outside her corporate email. Google said last month that it was investigating why Mitchell downloaded a large number of files and shared them with people outside the company.
- According to a source, Mitchell used automated texts to look at her messages to find examples showing discriminatory treatment of Gebru before her account was closed.
- The company on Friday refused to offer an update on Mitchell’s status, but from now on she will remain an employee of the company.
“Research leadership has talked a big game about accountability,” said ethical AI researcher Alex Hanna said in a tweet on Thursday. “But there is no liability for anything that has happened in the last three months. As far as I am concerned, research leadership does not know the meaning of the word.”